[CIG-SEISMO] blow up using Gaussian signal

[Yingzi Ying]

Hi all,

I am here just to report some unstable cases I met.

I simulated a wave field excited by a vertical force with Gaussian 
signal in elastic domain with SPECFEM3D. The running blew up with error 
message "forward simulation became unstable and blew up".

Similarly, I performed SH-wave simulation in SPECFEM2D using Gaussian 
signal, it blew up again. I attached the source signal and a received 
signal from a station in this simulation for your reference.

I tested other waveforms such as Ricker wavelet, they all went well. But 
when I changed source signal to Gaussian, it blew up. Different "hdur" 
and "dt" were also tested.

Best,
Yingzi

-------------- next part --------------
A non-text attachment was scrubbed...
Name: sourceSignal.eps
Type: application/postscript
Size: 39393 bytes
Desc: not available
Url : http://geodynamics.org/pipermail/cig-seismo/attachments/20130205/0375a817/attachment-0002.eps 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: receiverSignal.eps
Type: application/postscript
Size: 39862 bytes
Desc: not available
Url : http://geodynamics.org/pipermail/cig-seismo/attachments/20130205/0375a817/attachment-0003.eps